PROMERYCOCHCERUS 121 



Foramina: The infraorbital foramina lie above the anterior part of P 4 . The supraorbitals are 

 20 mm. apart and lie near the posterior part of the nasal bones. The foramen ovale is large. The 

 posterior palatine foramina are not apparent. 



Dentition: Superior: The metastyle of M 3 is medium-sized. There are internal cingula on 

 the molars and on P 4 . P 1 is isolated. The incisors are very small, and the canines are large. 

 Inferior: The metastylid of M 3 is very narrow and well offset from the hypoconid. The pre- 

 molars of both series are simple but typical of the genus. On P 3 the posterointermediate crest takes 

 off from the primary cusp, as usual. The superior molar-premolar index of the holotype is 0.83 and 

 of the paratype 0.76, while the inferior index of the holotype is 0.79. 



Skeleton: This differs from the normal Promerycochcerus form in that it has a short body and 

 long legs. There are thirteen dorsals, instead of fourteen as in P. carrikeri. 



Loomis (1933) has described the skeleton, Cat. No. 31-104 A.M. and the following charac- 

 ters are taken from his paper. The scapula is relatively long and narrow, with the spine well 

 developed. The long, light humerus is shorter than in Mesoreodon. The radius is long and 

 slender, while the olecranon process of the ulna is less developed than in any other oreodont. The 

 carpus is high, and the magnum is in contact with the unciform for more than half its height, a con- 

 dition similar to that found in Mesoreodon and Tic/ioleptus, as well as in Promerycochcerus. The 

 scaphoid has the greatest development so far known in any oreodont and occupies about half of the 

 proximal carpal row, in consequence of which the cuneiform is much reduced in size. In the distal 

 row the trapezoid is very much enlarged, more so than in any other oreodont, the nearest approach 

 to it in size being found in Ticholeptus. 



The long metacarpals are strongly in contrast with the normal stubby ones of Promerycochcerus. 

 The phalanges strongly resemble those of Mesoreodon and again contrast with the short toes of 

 Promerycochcerus. 



The pelvis, both in length and weight, compares well with that of Mesoreodon, Ticholeptus, 

 or Merycoidodon. The ischial and pubic portions are elongate and slender, but in Promerycochcerus 

 these parts are short, with strong tuberosities. The long, slender femur, tibia, and fibula all com- 

 pare well with those of Mesoreodon or Ticholeptus but are perhaps even more elongate than in 

 those genera. The tarsals and metatarsals are elongate, in marked contrast with the shortened ones 

 of the Promerycochcerus type. 



Discussion: The holotype is the basis for the above specific characters. The specimen is a 

 female, while the paratype skull is that of a male, according to Loomis in both instances. The para- 

 type differs somewhat from the holotype. 



It was the discovery of a specimen of this species in the holotype locality, possessing a short- 

 bodied, long-limbed skeleton, which led Loomis (1933) to erect the genus Hypselochcerus for its 

 reception. Subsequent discoveries by Schlaikjer show that the known skeletons of Promerycochcerus 

 grade into P. gregoryi, and it is possible that the shorter forms, with longer limbs, may be those of 

 females. 



The holotype of P. gregoryi and that of P. curvidens are exceedingly close to each other in 

 general characters. If Loomis' species should warrant a subgeneric rank on further evidence, then I 

 believe that it should be placed in my Desmatochcerus, which was proposed twelve years earlier, 

 rather than in the later Hypselochcerus. It is true that there is no skeleton of Desmatochcerus 

 known, but the type skulls are certainly no more than specifically distinct. 



The validity of Loomis' reference to P. gregoryi of the skeleton found at a later date has been 

 questioned. It is true that the accompanying skull differs in several respects from that of the holo- 

 type, but it seems to me that these differences are not of sufficient importance to invalidate the 

 reference. 



In conclusion, it is my belief that P. gregoryi, P. g. loomisi, and P. curvidens form a natural 

 association which, if taken out of Promerycochcerus, should be placed under Desmatochcerus but 

 which, on the present evidence, do not seem to warrant such a procedure, although it is quite 

 possible that they may be so separated in the future. 



